The CHiME challenge series aims to advance far field speech recognition technology by
promoting research at the interface of signal processing and automatic speech recognition …

This paper presents the design and outcomes of the CHiME-3 challenge, the first open
speech recognition evaluation designed to target the increasingly relevant multichannel …

Recently, frequency domain all-neural beamforming methods have achieved remarkable
progress for multichannel speech separation. In parallel, the integration of time domain …

In this paper, we present an optimal multi-channel Wiener filter, which consists of an
eigenvector beamformer and a single-channel postfilter. We show that both components …

Time-frequency (TF) masks are widely used in speech enhancement (SE). However,
accurately estimating TF masks from noisy speech remains a challenge to both statistical or …

The CHiME challenge series has been aiming to advance the development of robust
automatic speech recognition for use in everyday environments by encouraging research at …

Many mobile devices now include an extra microphone, frequently placed at their rear,
intended to obtain information about the environmental noise for speech de-noising …

Automatic Speech Recognition (ASR) usually works well with close-talking microphone
environment rather than in far-field conditions. A major challenge in the far-field ASR …

In the project an automatic speech system is used in mobile customer care services. In
existing mobile customer care services, customer have to wait for 4 to 5 minutes to get into …

Microphone array speech separation in high reverberant and low signal-to-noise ratio (SNR)
environments, is still a challenge. This paper utilizes a deep neural network (DNN) and …